NEXUSCode
Taking on new automation projects

Automation that runs your business.

Nexus Code builds AI-powered workflows, internal tools, and cloud systems for operations-heavy companies — then keeps them running.

30
Production systems
10+
Claude workflows live
~60K
Lines in production
1
Engineer, end to end

Services

The systems behind the work your team does every day

AI workflow automation

Claude wired into the work your team does by hand today: auditing, classification, OCR on handwritten notes, and written briefings.

Internal tools & portals

Dashboards for your staff and secure portals for your customers, with real authentication and the numbers they actually ask about.

Data & reporting

Revenue, commissions, and KPIs pulled together from the systems you already run, delivered on a schedule instead of on request.

Compliance & inspections

Digital, offline-capable forms with countersignatures and integrity checks, so audits are a report, not a scramble.

Cloud infrastructure

Serverless AWS, CI/CD with safe-deploy gates, and drift detection so nothing changes in production without someone knowing.

Security & reliability

Least-privilege access, hardened APIs, credential rotation, and runbooks — the unglamorous work that keeps systems up.

Featured work

Running a multi-shop operation on automation

Multi-shop heavy-truck repair & fleet services

Iron Buffalo Holdings

A 30-system automation platform, built and run by one engineer

Iron Buffalo needed reporting, compliance, commissions, and customer portals that kept up with a growing multi-shop operation. Nexus Code designed, shipped, and now operates the whole platform.

30
Systems
10+
AI workflows
45
Entities via MCP
Claude APIMCPAWS LambdaECS FargateSQL ServerApps ScriptGitHub Actions
  • Claude-powered invoice auditing, service-request classification, and handwritten-note OCR
  • Twice-daily AI-written operations briefings for leadership
  • Read-only MCP server so the team can ask plain-English questions of production data
  • DOT annual (49 CFR 396.17) and PM inspections with integrity checks that catch skipped work
  • Email-OTP customer portal serving fleet-readiness and spend KPIs to a national fleet account
  • GitHub Actions deploy gates plus daily drift and heartbeat detection

How it works

Small releases, real data, no big reveal

  1. 01

    Map the work

    We walk through how the process runs today — the spreadsheets, the inboxes, the copy-paste — and pick the piece that pays back fastest.

  2. 02

    Build it in production

    Small, working releases on your real data, with tests, access controls, and a deploy pipeline from day one. No six-month big reveal.

  3. 03

    Run it with you

    Monitoring, drift detection, and runbooks so it keeps working after launch. I stay on the hook for what I ship.

Who you work with

One engineer, from first call to on-call

No account managers and no hand-offs. The person who scopes your project is the person who builds it, deploys it, and answers when something needs attention.

Omar Daghestani

Founder & Engineer

Denver, CO

AI and automation engineer who designs, ships, and operates production systems end to end. Also an Infrastructure Support Engineer at Cognizant/TriZetto, and previously a full-stack teaching assistant at edX, where I mentored 200+ engineers.

  • B.S. Software Engineering, WGU
  • AWS Certified Cloud Practitioner
  • CompTIA Security+
PortfolioGitHubLinkedIn

Have a process that should run itself?

Tell me how it works today. You'll get a straight answer on what's worth automating first — usually within a day.

omar@nexuscodelabs.dev